Calorimetric determination of the heats of...

U.S. Department of Commerce Bureau of Standards

RESEARCH PAPER RP686

Part of Bureau of Standards Journal of Research, vol. 12, June 1934

CALORIMETRIC DETERMINATION OF THE HEATS OFCOMBUSTION OF ETHANE, PROPANE, NORMALBUTANE, AND NORMAL PENTANE

By Frederick D. Rossini

abstract

The present investigation on the calorimetric determination of the heats of

combustion of ethane, propane, normal butane, and normal pentane was under-taken because of the dearth of information concerning these thermochemicalconstants for which there is pressing need in industry and in the sciences.

The data of the present investigation give for the heats evolved in the com-bustion in oxygen, at a temperature of 25 C and a constant pressure of 1 atmos-phere, of gaseous ethane, gaseous propane, gaseous normal butane, and gaseousnormal pentane, respectively, to form gaseous carbon dioxide and liquid water,the following values, in international kilojoules per mole: Ethane, 1,559.57 ±0.44;propane, 2,219.57±0.51; normal butane, 2,877.88±0.63; normal pentane, 3,536.00±0.88. Converted to k-cali5 by means of the factor 1.00040/4.1850, these valuesbecome, in k-cali 5 per mole: Ethane, 372.81 ±0.11; propane, 530.57 ±0.12;normal butane, 687.94 ±0.15; normal pentane, 845.27 ±0.21.The hitherto "best" values for the heats of combustion of these gases differ

from the values obtained in the present investigation by the following amountsEthane, —1.23 percent; propane, —0.87 percent; normal butane, no existing

data; normal pentane, —0.86 percent. These differences are from 30 to 50times the estimated uncertainties in the values from the present investigation.

CONTENTSPage

I. Introduction 736II. Units, atomic weights, etc 736

III. Method and apparatus 737IV. Chemical procedure 739

1. Preparation and purity of the hydrocarbons 7392. Purity of the combustion reaction 7413. Determination of the amount of reaction 742

V. Calorimetric procedure 7421. Electrical energy experiments 7422. Correction experiments 7433. Reaction experiments 744

VI. Results of the present investigation 746

VII. Data of previous investigations 7471. Ethane 747

2. Propane 7473. Normal butane 748

4. Normal pentane 748VIII. Summary 749

IX. References 750

735

736 Bureau of Standards Journal of Research [Vol. n

I. INTRODUCTION

Accurate values for the heats of combustion of ethane, propane,normal butane, and normal pentane are needed for the followingreasons: First, ethane, propane, and butane are becoming increasingly

important as fuel gases in industry and in the home; second, for these

substances, values of the free energies of formation, which are neededfor computing chemical equilibria in which these gases take part, canbe determined accurately only from the relation involving entropiesand heats of formation, which latter values are calculated from theheats of combustion ; and third, accurate values for the heats of com-bustion of the normal paraffin hydrocarbon gases will give data fromwhich may be drawn information concerning the energies of theatomic linkages in these molecules.The existing data on the heats of combustion of ethane, propane,

butane, and pentane are very meager and, according to presentstandards, exceedingly uncertain. For ethane and propane there

are the discordant values of Thomsen (l) 1 and Berthelot and Matignon(2), who used the calorimetric equipment available half a centuryago and who had not pure gases to work with. For normal butane,there are no data on the heat of combustion recorded in the litera-

ture. For normal pentane there exists the value obtained by Rothand Machlett (3), who published no report of their experiments butgave simply the numerical results.

Because of the dearth of information concerning the heats of com-bustion of these gases and because of the pressing need in industryand in the sciences for accurate values concerning them, the presentinvestigation on the calorimetric determination of the heats of com-bustion of ethane, propane, normal butane, and normal pentane wasundertaken.

II. UNITS, ATOMIC WEIGHTS, ETC.

The fundamental unit of energy employed in the present calori-

metric experiments is the international joule based upon standards of

emf and resistance maintained at this Bureau (4). Recent experi-

mental work in this Bureau (13) on the relation between the interna-

tional and the absolute units of current and resistance indicates therelation between the absolute joule and the international joule,

derived from standards maintained at this Bureau, to be 1 interna-tional (Bureau of Standards) joule = 1.00032 absolute joules. Butfor the sake of consistency the relation which has heretofore been usedfor converting the data obtained in the thermochemical laboratoryof the Bureau of Standards will be used in the present and succeedingpapers, until such time as the relation between the absolute and theinternational joule will be fixed by international accord. At thattime, all the data that will have been obtained in this thermochemicallaboratory will be converted to the new basis.

The conversion factors used throughout this paper are:

1 international joule = 1.00040 absolute joules

1 g-calis = 4.1850 absolute joules

1 The numbers in parentheses, here and throughout the text, refer to the references, p. 750.

Rossini] Heats of Combustion 737

The atomic weights of oxygen, hydrogen, and carbon are takenfrom the 1934 report of the International Committee on AtomicWeights (5): Oxygen, 16.0000; hydrogen, 1.0078; carbon, 12.00. Asexplained in section IV, part 3, of this paper, the thermal data ob-tained in the present investigation are independent of the atomicweight of carbon, the number of moles of hydrocarbon involved in

the calorimetric experiments having been determined from the massof water formed.

III. METHOD AND APPARATUS

The general calorimetric method employed in the present investiga-

tion is the same as that previously used in this laboratory in thedetermination of the heats of formation of water and hydrogen chloride

and the heats of combustion of methane, carbon monoxide, methanol,and ethanol (6); and is, briefly, a substitution method in which theheat evolved by a measured amount of the chemical reaction is

found to be equivalent to the heat evolved by a measured amount of

electrical energy, the calorimeter serving as the comparator. Unlessotherwise stated the procedure and the nomenclature follow theprevious investigations (6).

For the combustion of ethane, it was found that the reaction vessel

used for the combustion of methane served very well, provided thatthe proper conditions prevailed with respect to the rates of flow of theethane and the oxygen. This reaction vessel has already beendescribed (6).

For the combustion of propane, normal butane, and normal pentanea new reaction vessel was designed, in which provision was made for

aspirating oxygen into the stream of hydrocarbon gas before it

reached the burner tip.2 A cross section of this new reaction vessel 3

is shown in figure 1. A is the tube through which the oxygen enters

the reaction chamber; B is the tube through which the hydrocarbongas passes to the burner tip F where, having been ignited by sparksacross the gap between the ends of the platinum wires D, it burns witha flame; at E is the injector by means of which oxygen is drawn into

the stream of hydrocarbon gas; at G is collected as a liquid most of

the water formed in the combustion; and the exit gases leave thereaction vessel through the tube C.The dimensions of the new reaction vessel necessitated the building

of a new heating coil to fit it. A thin copper cylinder, 5 cm long and7.5 cm in diameter, was coated on the outside with a layer of pizein

cement; around this cylinder was wound, noninductively, about 65ohms of enameled constantan wire (no. 30, B. & 3. gage); enameledcopper leads were soldered to the ends of the constantan wire andinsulated with pizein on the cylinder; and the outside of the cylinder

was covered with another coating of pizein.

2 The author is indebted to Francis A. Smith for valuable advice in the design of the burner.3 This reaction vessel was built by E. O. Sperling, and the author is indebted to him for improvements in

the design.

738 Bureau oj Standards Journal of Research Vol. 12

1*1 ft

Figure 1.

—

Crosssection of the reaction vessel used in the combustion of propane,

butane, and pentane.

The description is given in the text. The drawing is made to scale, and the over-all length is 26 cm.


IV. CHEMICAL PROCEDURE1. PREPARATION AND PURITY OF THE HYDROCARBONS

In the present experiments the amount of reaction was determinedfrom the mass of water formed. Because the heats of combustion of

these hydrocarbons per mole of water formed are not greatly different,

quite an appreciable amount of a homologous impurity may be con-tained in a given sample without necessitating any significant cor-

rection to the observed heat effect. For example, if a sample of

normal butane contains as impurity 0.2 mole percent of propane,the observed heat effect will be less than that for pure normal butaneby 0.006 percent. Table 1 gives the amounts of various impuritiesnecessary to cause an error of ± 0.01 percent in the heat of combus-tion.

Table 1.— The effect of impurities on the heat of combustion

Substance Impurity

Amount of im-purity neces-sary to causean error of 0.01

percent in theheat of com-

bustion

Error inthe heat ofcombus-

tion

Ethane

Propane

Normal butane-

Normal pentane

(Methane1 Propane.I Ethane.

-

1 Butane..fPropane.\ Pentane./Butane. .

\Hexane_.

Mole percent0.10.11.21.21.35.35.57.57

Percent-0.01+.01-.01+.01-.01+.01-.01+.01

The samples of ethane, propane, and normal butane used in thepresent investigation were prepared to our order by the laboratoryof the Linde Air Products Co., Buffalo, N.Y. These saturated hydro-carbons were obtained from natural gas by repeated fractional dis-

tillation at low temperatures. The manner of purification was suchthat each gas was expected to contain less than 0.2 percent by volumeof impurity, which impurity would be a homologous substance, as

for example, an impurity of ethane or butane in the sample of pro-

pane. Subsequent tests made in our laboratory indicated that the

amount of impurity in the various samples was, in general, consider-

ably less than 0.2 percent. About 2.4 moles of each gas was placed in

highly evacuated steel cylinders, which were fitted with good valves

and had internal volumes of about 2.8 liters.

In our laboratory, the ethane, propane, and normal butane wereexamined for impurities by Martin Shepherd (7), who separated,

by isothermal distillation at low pressures, a sample of each gas into

three fractions, and measured the difference in the vapor pres-

sures of the middle cut and the initial distillate and of the mid-dle cut and the final residue. The amounts of the initial distillate

and the final residue were each less than 1/20 of the entire sample,

and the difference in vapor pressures was measured by a differential

manometer sensitive to ± 0.02 mm of Hg. The data obtained byShepherd are given in table 2, where Ap is the difference between the

vapor pressures, measured at the temperature t, of the middle cut

and the initial distillate or the final residue, and AXB ,(see (7) for

method of calculation and exact definition of terms) , represents approx-

740 Bureau oj Standards Journal of Research [Vol. is

Table 2.

—

Data on the purity of the ethane, propane, and normal butane, as shownby measurement of the difference in the vapor pressures between the middle cut andthe initial distillate and the final residue, respectively

Initial distillate ' Final residue 1

Substance Ap t AXb Ap t AXB

mm H<?

0.29.00

7.82

°C-189.

1

-58.2

.0

Mole fraction0.009 methane(0.00001 ethane

JO.003 propane1 0.020 isobutane

mmHg0.01.02

}.22

°C-150.0-49.3

.0

Mole fraction0.005 propane.0.00004 normal or isobutane.

Normal butane. __ 0.00035 normal pentane.

> The amounts of the initial distillate and the final residue were respectively less than 1/20 of the entire

sample. The value of AXs gives approximately the mole fraction of the given impurity in the initial dis-

tillate or the final residue, not in the entire sample. The values of Ap were read to ±0.02 mm Hg.

imately the mole fraction of a possible impurity, B, present in the initial

distillate or the final residue. The mole fraction of the impurity B in

the entire sample would be about 1/20 of the value of AXB given in

table 2, were it not for the fact that the isothermal distillation doesnot isolate all of the impurity in the initial distillate or the final resi-

due. Assuming, as a probable case, that one half of the impurity is

obtained in the initial distillate or the final residue, then the molefraction of the impurity B in the entire sample would be about 1/10

the value of AXB .

The normal pentane used in the present experiments was from the

lot prepared synthetically by Mair (8), who gave a complete reportof its preparation and properties. The following are some of thephysical constants of the normal pentane: boiling point at 760 mmHg, 36.06 degrees C; freezing point in air, — 129.73 degrees C; freezing

range of the sample, 0.026° C; refractive index, ND25

, 1.35470.

Table 3.

—

Results of the analytical experiments for determining the ratio of carbonto hydrogen in the ethane, propane, normal butane, and normal pentane

SubstanceNumberof experi-ments

Averageamount ofhydrocar-bon usedin each ex-

periment

RatioAveragevalue of

the ratio

Averagedeviationfrom themean

Maximumdeviationfrom themean

Ethane ._..__ 5

7

6

6

Mole0.049.034.025.019

6C/2H8C/3H10C/4H12C/5H

0. 999921. 000161. 00014

. 99994

± 0. C0018± . 00019± . 00012± . 00022

0. 00039Propane . 00034

. 00028

. 00041

In table 3 are given the results of experiments on the determinationof the ratio of carbon to hydrogen in the ethane, propane, normalbutane, and normal pentane. In these experiments the combustionwas carried out in the calorimetric reaction vessel and the analyticalprocedure was the same as that used for methane (6). The amountof carbon dioxide collected in these experiments was corrected, whennecessary, for the small amounts of carbon monoxide formed. (See

p. 741.)

From the foregoing data, it appears that the purity of the ethane,propane, normal butane, and normal pentane used in the present in-

vestigation was such that, per mole of water formed, the heats ofcombustion of these samples would differ from the heats of combustionof the absolutely pure hydrocarbons by less than 0.01 percent.


2. PURITY OF THE COMBUSTION REACTION

The purity of the combustion reaction was examined as reportedin previous papers.A new supply of oxygen, prepared commercially from liquid air,

was obtained for the present experiments. This oxygen, like thatpreviously used, was low in inert gases (0.6 mole percent) and wassimilarly purified. A flow of oxygen considerably in excess of thestoichiometric requirements for complete combustion was used in all

the experiments.The combustion of ethane was carried out in the reaction vessel

used for methane and other gases (6). It was necessary to use sucha rate of flow of the ethane as would give a flame with a blue mantle,a sign of practically complete combustion, and which was sufficiently

removed from the burner tip to prevent decomposition (with conse-quent deposition of carbon on the quartz tip) of the ethane before it

reached the flame proper. The desired ethane flame was about 1.5

cm long. In the combustion of the ethane under these conditions,

it was found that, on the average, 0.008 percent of the carbon in thesample was burned to carbon monoxide instead of carbon dioxide.

This necessitated a correction of + 0.003 percent to the observedheat effect in the calorimetric reaction experiments, and of + 0.008percent to the observed mass of carbon dioxide in the analytical ex-

periments on the ratio of carbon to hydrogen in the ethane.For the combustion of the propane and normal butane, the reaction

vessel shown in figure 1 was used. The aspiration of oxygen into

the stream of hydrocarbon gas before it reached the flame madepossible the combustion of these hydrocarbons without depositionof carbon on the burner tip. The amount of oxygen aspirated into

the stream of hydrocarbon gas was kept below that amount whichwould cause the flame to flash back to the capillary jet, by governingthe size of the injector holes and the rate of flow of the hydrocarbongas. In the combustion experiments with propane and normal butaneburned in this manner, with a blue flame about 1.5 cm long, no carbonmonoxide was detected in the exit gases.

The combustion of the normal pentane, which was carried as a

gas into the reaction vessel by means of a stream of nitrogen, wasperformed in the same reaction vessel that was used for propaneand butane. The composition of the gaseous mixture of nitrogen

and pentane was about 0.65 mole of normal pentane to 0.35 mole of

nitrogen. The nitrogen was bubbled through the liquid normal pen-tane at a temperature of 24 C and then the mixture passed throughascarite, magnesium perchlorate, and phosphorus pentoxide before

arriving at the calorimeter. The pentane flame was blue and aboutthe same size as for the other gases. The amount of carbonmonoxide- formed in the combustion of the normal pentane in this

manner was found to be 0.00020 mole per mole of carbon dioxide,

which necessitated a correction of + 0.008 percent to the observedheat effect in the calorimetric reaction experiments, and a correction of

+ 0.020 percent to the observed mass of carbon dioxide in the ana-lytical experiments on the ratio of carbon to hydrogen. The amountof nitrogen oxides in the exit gases from the combustion of pentane

* The tests for carbon monoxide were made in the gas section by Carroll Creitz.

742 Bureau of Standards Journal of Research \voi.12

was found to be 0.00004 mole per mole of carbon dioxide,5 whichnecessitated no significant correction.

3. DETERMINATION OF THE AMOUNT OF REACTION

The amount of reaction in each calorimetric combustion experi-

ment was computed from the mass of water formed, which was deter-

mined according to our usual procedure (6). In this connection it

should be emphasized that for work of the highest precision the

U-tube absorbers should be filled with hydrogen at the times of

weighing, in order to make negligible any errors in corrections dueto changes in the room temperature or the barometric pressure, andto make more certain the correction for the change in the volume of

the solid absorbent.

In the piesent experiments 1 mole (18.0156 g) of H2 was takenas equivalent to 1/3 mole of C2H6 , to 1/4 mole of C3H8 , to 1/5 mole of

C4H10, or to 1/6 mole of C 5Hi2 . In this maimer the values of the

heats of combustion obtained in the present investigation are inde-

pendent of the atomic weight of carbon.

In an auxiliary experiment an amount of the gaseous mixture of

normal pentane and nitrogen containing 0.17 g of pentane was passedthrough a carefully weighed U-tube containing dehydrite (magnesiumperchlorate trihydrate) and phosphorus pentoxide, and the changein mass of the U-tube was found to be zero ± 0.00005 g. Thisdetermination was necessary because, at the extinction of the pentaneflame, there was always several mg of unburned pentane in the inlet

tube, which pentane was later to be swept out through the U-tube.

V. CALORIMETRIC PROCEDURE1. ELECTRICAL ENERGY EXPERIMENTS

The data of the experiments made to determine the electrical

energy equivalent of the calorimeter system used for the combustionof ethane are given in table 4. (See (6) for nomenclature.) The"error" of the mean value of these experiments is

± 2-yj2A2/n(n - 1) = ± 0.012 percent,

where S A 2is the sum of the squares of the percentage deviations

and n is the number of experiments. 6

The data of the experiments made to determine the electrical

energy equivalent of the calorimeter system used in the combustionof propane, normal butane, and normal pentane are given in table 5.

The "error" of the mean value of these experiments is

± 2-yfXA2/n{n - 1) = ± 0.008 percent

* The test for nitrogen oxides was made in the reagents and platinum metals section by F. W. Schwab.6 This formula gives an "error" that is three times as large as that obtained with the usual formula for

the "probable error" of the mean.


Table 4.

—

Calorimetric results of the electrical energy experiments 1 for ethane

Experiment no. Aft k K U A^corr.

Aver-agetem-pera-ture

Electri-

cal

energy 2

Mass of

calori-

meterwater

Electri-

cal

energyequiv-alent of

calori-

metersystem 3

Devi-ationfrommean

1

Ohm0. 298843

. 289622

. 289200

. 294777

. 296576

. 299513

. 297286

. 297230

. 296837

Min- 1

0. 001969. 001943. 001937. 001948. 001933. 001946. 001942. 001935. 001931

Ohm0. 005401

. 005885

. 005867

. 005835

. 005938

. 005366

. 005430

. 005355

. 005367

Ohm0. 000254.000110. 000046

-.000140. 000012

-. 000014. 000007. 000048. 000226

DegreesC

2. 906362.811562. 808192. 865662. 880972. 916012. 893092. 892872. 88709

DegreesC25.1025.0625.0625.0925.0925.1025.0925.1025.10

Int.

joules

44, 944. 5

43, 484. 2

43, 392. 7

44, 257. 4

44, 539. 5

45, 038. 2

44, 673. 9

44, 655. 2

44, 563. 2

Q3, 625. 40

3, 624. 30

3, 623. 14

3, 621. 44

3, 624. 94

3, 620. 49

3, 620. 69

3, 617. 93

3, 619. 94

Int.

joulesper de-

gree C15,441.815, 441. 2

15, 439. 215, 438. 3

15, 439. 4

15, 443. 215, 438. 8

15, 445.

1

15, 435. 7

Joulesper

degreeC1.5

2 .9

3 -1.14 . -2.05... -.96 -_ 2.97 -1.58 4.89. . -4.6

15, 440. 3 ±2.2

i These experiments were performed in January 1933.2 The time of electrical energy input was 17 or 18 minutes, the current 0.78 to 0.83 ampere.3 Corrected to 3,620.00 g of water and an average temperature of 25.00 C.

Table 5.

—

Calorimetric results of the electrical energy experiments 1 for propane,butane, and pentane

Experiment no. Aft k K U Atcorr.

Aver-agetem-pera-ture

Electri-

cal

energy 2

Mass of

calori-

meterwater

Electri-

cal

energyequiv-alent of

calori-

metersystem 3

Devi-ationfrommean

1 . _

Ohm0. 410460

. 408615

. 412305

. 411657

. 411322

. 413270

. 409501

. 410563

. 409252

. 408933

. 410085

. 407731

Min" 1

0. 001932. 001934. 001904. 001940. 001930. 001925. 001934. 001934. 001926. 001941. 001941. 001943

Ohm0. 007159

. 007255

. 007074

. 007156

. 007245

. 007277

. 006995

. 006721

. 006818

. 007053

. 006848

. 006930

Ohm0. 000156

. 000166-. 000034

. 000084

. 000075

. 000053-. 000046-. 000223

. 000000-. 0G0093-. 000055-. 000190

DegreesC

3. 996283. 976954. 017304. 008934. 004794. 024003. 990404. 005403. 989233. 984663. 997743. 97493

DegreesC25.0125.0025.0125.0325.0125.0225.0025. 0025.0025.0025.0025.00

Int.

joules

60, 887. 8

60, 962.

61, 097. 8

60, 991.

1

60, 950. 2

61, 236.

60, 814. 1

60, 880.

60, 705. 9

60, 673. 4

60, 830. 3

60, 454. 7

Q3, 561. 77

3, 583. 62

3, 554. 76

3, 555. 93

3, 557. 44

3, 557. 87

3, 563. 15

3, 552. 023, 556. 923, 559. 35

3, 555. 97

3, 554. 44

Int.

joulesper de-

gree C15, 228. 7

15, 230. 215, 230. 6

15, 230. 8

15, 230.

15, 226. 6

15, 226. 9

15, 232. 8

15, 230. 3

15, 229. 4

15, 233.

15, 232. 2

Joulesper

degreeC— 1.4

2 . 1

3 .54 .7

7-".--"""""

"

-.1-3.5—3.2

8 2.79 .210- . — 7

11_._ 2.912 2. 1

Mean 15, 230.

1

±1.5

1 Experiments 1 to 6 were performed in September 1933, and experiments 7 to 12 in January 1934.2 The time of electrical energy input was 16 or 17 minutes, the current 0.95 to 1.00 ampere.3 Corrected to 3,560.00 g of water and an average temperature of 25.00 C.

2. CORRECTION EXPERIMENTS

The " spark" energy was determined calorimetrically to be 1.43

± 0.06 joules per second for the experiments on ethane and 1.31 ±0.06 joules per second for the experiments on propane, normal butane,and normal pentane.

55948—34-

744 Bureau of Standards Journal of Research [Vol. IS

The "gas" energy and the "vaporization" energy were deter-

mined as in the previous investigations (6).

As before (6), calorimetric experiments were performed in whicha number of ignitions and extinctions of the flame occurred, thepurpose being to subordinate, relatively, the heat evolved in the

combustion of the hydrocarbon to the "spark", "gas", and "vapor-ization" energies. In each of these experiments thermal balancewas obtained within the limits of error of the measurements.

3. REACTION EXPERIMENTS

The calorimetric data of the combustion experiments with ethane,

propane, normal butane, and normal pentane are given in tables

6, 7, 8, and 9, respectively. The "errors" of the mean values of

the combustion experiments on the various gases, computed as

± 2^2A 2jn(n - 1), are, respectively: Ethane, ± 0.023 percent; pro-

pane, ± 0.019 percent; normal butane, ± 0.018 percent; and normalpentane, ± 0.021 percent.

Table 6.

—

Calorimetric results of the reaction experiments for ethane

Experiment no. AR k K U A tcOIT.

Averagetempera-

ture

1

Ohm0. 290372

. 294673

. 296344

. 298509

. 295834

. 296406

. 294781

0. 001898. 001942. 001923. 001933. 001980. 001957. 001945

Ohm0. 006838

. 005332

. 005790

. 005946

. 006425

. 006635

. 005662

Ohm-0. 000069

. 000137

. 000020-. 000156

. 000091-. 000075-. 000050

Degrees C2. 811352. 866872. 880052.901712. 868002. 873232. 86652

Degrees C25.09

2.. 25.083 25.084 25.095 25.086 . 25.087 25.08

Mean ..

Experiment no.

Electricalenergy

equivalent of

calorimetersystem i

"Gas"energy

"Spark"energy

"Vapori-zation"energy

Mass ofethane

Heat 2 of

combus-tion at

25degreesC

Deviationfrommean

1

Int. joulesper degree

C15, 442.

1

15, 459. 3

15, 451.

1

15, 455.

15, 456. 9

15, 455.

15, 430. 7

Joules4.613.98.613.95.912.413.1

Joules17.214.314.315.022.932.915.7

Joules375.8342.9303.4298.0331.7326.8261.9

Mole0. 0280855

. 0286407

. 0287294

. 0289423

. 0286226

. 0286718

. 0285167

Int. kilo-

joules permole1558. 74

1559. 461559. 351559. 81

1559. 841559. 51

1560. 26

Kilojouleper mole

-0.832 -. 11

3 -.224 .245 .276 -.061.. . .69

Mean 1559. 57 ± .35

1 Includes the heat capacity of H the mass of liquid water formed in the reaction.2 Includes a correction of +0.003 percent for the carbon monoxide formed.

Rossini] Heats oj Combustion 745

Table 7.

—

Calorimetric results of the reaction experiments for propane

Experiment no. AR k K U A Icorr.

Averagetempera-

ture

1

Ohm0. 403093

. 412941

.411032

. 413092

. 409675

. 406890

. 412252. 409171

Min-i0. 001956

. 001940

. 001922

. 001935

. 001947

. 001936

. 001948

. 001945

Ohm0.011899.015120. 007486. 008225. 007578. 007383. 010572. 008632

Ohm-0. 000179-.000008-.000011-.000008-.000081.000098. 000027.000059

Degrees C3. 879553. 943594.000374. 013433. 986703. 959253. 981493. 96987

Degrees C24.97

2 25.02

3 25.014 . ___ - 25.025 ... 25.01

6 24.9925.01

8 25.00

Experiment no.

Electrical

energyequivalent of

calorimetersystem i

"Gas"energy

"Spark"energy


Mass of

propane

Heat of

combus-tion at

25

degreesC

Deviationfrommean


C15, 197. 4

15, 235. 8

15,231.215, 230. 815, 223. 6

15, 225. 4

15, 232. 4

15, 230. 9

Joules-4.0-8.34.86.9

-2.66.32.9.6

Joules57.66S.224.932.713.123.635.419.7

Joules451.2340.5260.5294.6278.0280.5281.2257.8

Mole0. 0257306.0271917. 0275458. 0276635. 0274649. 0272818. 0274421. 0273537

Int. kilo-


2219. 342220. 692219. 402219. 352219. 222219. 082219. 20

Kilojoulesper mole

0.682 -.233 1.124 _ -.175 -.226 -.357 -.498 -.37

Mean . __ 2219. 57 ±.45

• Includes the beat capacity of Vi tbe mass of liquid water formed in the reaction.

Table 8.

—

Calorimetric results of the reaction experiments for normal butane

Experiment no. AR k K U At cotr.

Averagetempera-

ture

1-.Ohm

0.409115. 410905. 409376.411469.411537. 409147. 411499. 410687

Min-t0. 001946

. 001939

. 001940

. 001925

. 001923

. 001927

. 001928

. 001935

Ohm0. 006179

. 006254

. 006779

. 006667

. 006314

. 006896

. 006312

. 006511

Ohm-0. 000048-.000024-.000074-.000053-. 000041-.000139-.000170-.000142

Degrees C3. 994694.011453. 991584. 013234. 017293. 988804. 018214. 00791

Degrees C25.00

2 25.003 25.004 25.005 _____ — 25.006.. 25.007 25.008 25.00

Mean. ....

Experiment no.

Electricalenergy

equivalent of

calorimetersystem '

"Gas"energy

"Spark"energy


Mass of

normalbutane

Heat of

combus-tion at

25

degreesC

Deviationfrommean

1 .. ..


C15, 254.

15, 260. 4

15, 208. 5

15, 224. 6

15, 250. 5

15, 248. 2

15, 218. 2

15, 222.

1

Joules6.94.07.44.65.11.92.42.4

Joules7.9

14.423.617.015.715.717.014.4

Joules221.7207.3236.3221.9220.2238.8221.5216.6

Mole0. 0212578

. 0213428

.0211713

. 0213016

. 0213528

. 0212144

. 0213140

. 0212752

Int. kilo-

joules permole2876. 902877. 492877. 802878. 17

2879 062877. 632878. 722877. 23

Kilojoulesper mole

-0.9823 .. ...

-.39—.08

4 .295

6__.1.18-.25.84

8 -.65

Mean. . _ 2877. 88 ±.58

i Includes the heat capacity of }$ the mass of liquid water formed in the reaction.

746 Bureau of Standards Journal oj Research [Vol. 12

Table 9.

—

Calorimetric results of the reaction experiments for normal pentane

Experiment no. &R k K U At corr.

Averagetempera-

ture

1

Ohm0. 410430

. 409*539

. 407770

. 409557

. 411987

. 409974

. 413547

. 410295

Min~ l

0. 001937. 001936. C01952. 001950. 001949. 001943. 001925. 001945

Ohm0. 005964

. 006180

. 007482

. 006136

. 005585

. 008808

. 006274

. 008162

Ohm-0. 000108-.000134-. 000213-.000106-.000127-. 000122-. 000046-.000094

Degrees C4. 010454. 000723. 970074. 000074. 029633. 997704. 037664. 00701

Degrees C25.0025.0025.0025.00

2

34

5

6 25 007 25 018. 25 01

Mean .. .. ....

Experiment no.

Electricalenergy

equivalent of

calorimetersystem i

"Gas"energy

"Spark"energy

"Vapori-zation '

'

energy

Mass of

normalpentane

Heat 2 of

combus-tion at

25degrees

C

Deviationfrommean

1


C15, 180. 4

15, 230. 2

15, 229. 6

15, 222. 7

15, 258. 6

15, 214. 3

15, 063. 6

15, 213. 7

Joules4.02.84.06.2

-1.11.7

-2.51.1

Joules9.8

21.614.417.015.124.922.318.3

Joules211.0251.2350.5229.0256.6282. 7

255. 6

240.5

Mole0. 0172503

. 0172938

. 0172008

. 0172761

. 0174528

. 0172768

. 0172734

.0173076

Int. kilo-


3537. 053535. 15

3537. 543537. 073535. 763534. 763535. 41

Kilojoulesper mole

—0.752... 1.053 —.854 1.545 1.076... —.247... -1.248 —.59

3536. 00 ±.92

1 Includes the heat capacity of H the mass of liquid water formed in the reaction.2 Includes a correction of + 0.008 percent for the carbon monoxide formed.

VI. RESULTS OF THE PRESENT INVESTIGATION

The uncertainties in the final values obtained from the presentdata have been computed by taking the total percentage " error"

as equal to ± V^i2 + e2

2+ ez2

, where e x is the percentage " error" of themean value of the calorimetric experiments with electrical energy,e2 is the percentage " error" of the mean value of the calorimetric

combustion experiments, and e3 is an "error" of 0.01 percent assumedin the determination of the amount of reaction, due to impurities andother causes.

The data of the present investigation then yield for the heats of

combustion, at a temperature of 25 C and a constant total pressure of 1

atmosphere, of gaseous ethane, gaseous propane, gaseous normalbutane, and gaseous normal pentane, respectively, with gaseousoxygen to form gaseous carbon dioxide and liquid water, the following

values, in international kilojoules per mole:

Ethane 1,559.57 ± 0.44Propane 2,219.57 ± .51

Normal butane 2,877.88 ± .63

Normal pentane. __ 3,536.00 ± .88

Rossini) Heats of Combustion 747

VII. DATA OF PREVIOUS INVESTIGATIONS

1. ETHANE

The existing data on the heat of combustion of ethane are those of

Thomsen (1), and Berthelot and Matignon (2).

Thomsen performed seven experiments on ethane in his flamecalorimeter at constant pressure and at an average temperature of

19 C. The first three experiments were made with a sample of ethanehaving the empirical formula C2H6 . 54, prepared from zinc ethyl andhydrochloric acid; the last four with a sample having the formulaC 2H6 .033, prepared by the electrolysis of sodium acetate. Thomsendetermined the amount of reaction from the mass of carbon dioxide

formed. The writer has recomputed Thomsen's data, including a

correction for an impurity of methane, and obtains from his work,for the heat of combustion of ethane at a temperature of 25 C and a

constant pressure of 1 atmosphere, the value 368.9 ±1.3 k-cali5 , or

1,543.2 ±5.5 international kilojoules, per mole.Berthelot and Matignon (2) performed three experiments on the

heat of combustion of ethane, using a bomb calorimeter and workingat an average temperature of 13 C. The amount of reaction wasdetermined from the mass of carbon dioxide formed. Their recom-puted data yield for the heat of combustion of ethane, at a tempera-ture of 25 C and a constant pressure of 1 atmosphere, the value372.1 ±0.8 k-caJ !5 , or 1,556.6 ±3.4 international kilojoules, per mole.The various data on the heat of combustion of ethane are com-

pared in figure 2. The values from Thomsen (1) and Berthelot andMatignon (2) are lower than the value from the present investigation

by 1.05 and 0.19 percent, respectively.

2. PROPANE

The existing data on the heat of combustion of propane are thoseof Thomsen (1) and Berthelot and Matignon (2).

Thomsen performed seven experiments in his flame calorimeter,

at constant pressure and at an average temperature of 17.8 C, on a

sample of propane having the empirical formula C3H7.991, preparedfrom isopropyl iodide, zinc, and hydrochloric acid. The amount of

reaction was determined from the mass of carbon dioxide formed.The recomputed data of Thomsen, including a correction for impurity,yield for the heat of combustion of propane, at a temperature of

of 25 C and a constant pressure of 1 atmosphere, the value 528.7±1.2 k-cali 5 , or 2,211.7 ± 5.0 international kilojoules, per mole.Berthelot and Matignon performed three experiments on propane

in their bomb calorimeter at an average temperature of 17 C. Theamount of reaction was determined from the mass of carbon dioxideformed. Their recomputed data yield for the heat of combustion of

propane, at a temperature of 25 C and a constant pressure of 1

atmosphere, the value of 527.9 ± 1.8 k-cali 5 , or 2208.4 ±7.5 interna-tional kilojoules, per mole.The various data on the heat of combustion of propane are com-

pared in figure 2. The values from Thomsen (1), and Berthelot andMatignon (2) are lower than the value from the present investigationby 0.35 and 0.50 percent, respectively.

748 Bureau of Standards Journal of Research

3. NORMAL BUTANE

[Vol. It

There are no data on the heat of combustion of normal butanerecorded in the literature.

4. NORMAL PENTANE

The only value for the heat of combustion of normal pentanerecorded in the literature is that of Koth and Machlett (3). These

PROPANE

1550

1540

2220BR

2210

— T

2200

NORMALPENTANE

3540

«»R

3520

3500

RM

Figuee 2.

—

Plot of the various data on the heats of combustion of ethane, propane,and normal pentane.

The ordinate scale gives the heat of combustion at a temperature of 25 C and a constant total pressure of

1 atmosphere in international kilojoules per mole. The average values of each investigation, togetherwith the estimated uncertainties, are indicated as follows: T, Thomsen; BM, Berthelot and Matignon;RM, Roth and Machlett; R. Rossini.

authors have not published a report of their experiments, and giveonly the numerical result. Roth and Machlett found for the heat of

combustion of liquid normal pentane, at constant volume in a bombcalorimeter at a temperature of 20 C, the value 11,551 g-cal 15 per gramof liquid normal pentane, weighed in air. Applying the appropriatecorrections, this value becomes 832.8 k-cali5 per true mole of liquid

normal pentane, at a temperature of 25 C and a constant pressure of

1 atmosphere (lack of information prevents making the Washburn


correction). Roth and Machlett give for the heat of vaporization of

liquid normal pentane at 20 degrees C the value 4.9 k-eal per mole.

This value is apparently too low, as the calorimetric data of Griffiths

and Awbery (9) yield 6.2 k-cal per mole for the heat of vaporization

at 25 degrees C, and the data on the variation of the vapor pressure

with temperature (10) yield 6.6 k-cal per mole. Taking the directly

measured value of 6.2, the data of Roth and Machlett yield for the

heat of combustion of gaseous normal pentane, at a temperature of

25 C and a constant total pressure of 1 atmosphere, the value 839.0

k-cali 5 , or 3509.8 international kilojoules, per mole. If the value for

the heat of vaporization given by Roth and Machlett were used,

the above value would be lowered by 1.3 k-cal, or 5.4 kilojoules.

The data on the heat of combustion of gaseous normal pentaneare compared in figure 2. The value from Roth and Machlett (3)

is lower than that from the present investigation by 0.74 percent.

VIII. SUMMARYThe results of the present investigation confirm the suspected

existence of exceedingly large uncertainties in the hitherto existing

"best" values for the heats of combustion of ethane, propane, normalbutane, and normal pentane. The values from the previous investiga-

tions given in the preceding section of the present paper were com-puted from the original data by this writer; while the values whichhave mostly been used for the heats of combustion of these gases

are those given in the International Critical Tables (11). These latter

values were taken from the compilation by Kharasch (12), and, whencorrected to a temperature of 25 C, are all lower than those from thepresent investigation by the following respective amounts: Ethane,1.23 percent; propane, 0.87 percent; normal butane, no existing data;normal pentane, 0.86 percent. These differences are from 30 to 50times the estimated uncertainties in the values obtained from thepresent investigation.

The data of the present investigation give, for the heat involved in

the reaction, C raH2w+2 (gas) H -— 2 (gas) =nC02 (gas) + (n + 1)H2

(liquid) at a temperature of 25 C and a constant total pressure of 1

atmosphere, the following values for ethane, propane, normal butane,and normal pentane:

Table 10.

—

Results of the present investigation

Formula State

Heat of combustion a at 25degrees C and a constant pres-sure of 1 atmosphere

SubstanceInternationalkilojoules permole (funda-mental unit)

k-calis per mole(defined unit)

Ethane.. C2H6 Gas 1559. 57±0. 442219. 57± . 51

2877. 88± . 633536. 00± . 88

372. 81±0. 11

Propane- CaHs- Gas.. 530. 57± . 12Normal butane - 7J-C4H10--- Gas 687. 94± . 15Normal pentane 71-C5H12 Gas 845. 27± . 21

"As explained in sec. II, p. 736, the fundamental unit of energy in the present investigation is the in-

ternational joule based upon standards maintained at this Bureau, while the k-cahs is a defined unit, 1 k-calisbeing taken equivalent to 4.1850/1.00040 international kilojoules. One mole (18.0156 g, true mass) of H2Ois taken as equivalent to J-S mole of C2H6, to \i mole of C3H8, to Yb mole of C4H10, and to Ve mole of C5H12.

750 Bureau of Standards Journal of Research [Vol. ti

Data have now been obtained in this laboratory on the heats of

combustion of all the normal paraffin hydrocarbons in the gaseousstate from methane to pentane, inclusive. In a subsequent paper,

these data will be utilized: First, in the calculation of the heats of

formation of these molecules; second, in the accurate prediction of theheats of combustion and of formation of the gaseous normal paraffin

hydrocarbons containing more than five carbon atoms in the molecule;and third, in drawing conclusions concerning the energies of the atomiclinkages in these molecules.

The author is deeply grateful to the late Dr. Edward W. Washburnfor his helpful advice, his many valued suggestions, and his keeninterest in this work.

IX. REFERENCES

1. Thomsen, J., Thermochemische Untersuchungen, vol. IV, pp. 50-54. BarthLeipzig, 1886.

2. Berthelot, M., and Matignon, C, Ann. chim. phys., vol. 30, p. 547, 1893.3. Roth, W. A., and Machlett, unpublished. The numerical result appears in

the Landolt-Bornstein-Roth-Scheel Tabellen, 5th ed.. vol. II, p. 1588.Springer, Berlin, 1923.

4. Vinal, G. W., B.S.Jour. Research, vol. 8, p. 448, 1932.5. Baxter, G. P., Curie, M., Honigschmid, O., LeBeau, P., and Meyer, R. J.,

J. Am. Chem. Soc, vol. 56, p. 753, 1934.6. Rossini, F. D., B.S.Jour. Research, vol. 6, pp. 1, 37, 1931; vol. 7, p. 329, 1931;

vol. 8, p. 119, 1932; vol. 9, p. 679, 1932.7. Shepherd, M., B.S.Jour. Research, vol. 12, p. 185, 1934.8. Mair, B. J., B.S.Jour. Research, vol. 9, p. 457, 1932.9. Griffiths, E., and Awbery, J. H., Engineering, vol. 133, p. 84, 1932.

10. International Critical Tables, vol. 3, p. 220. McGraw-Hill Book Companv,New York, 1928.

11. Ibid., vol. 5, p. 163.

12. Kharasch, M. S., B.S.Jour. Research, vol. 2, p. 359, 1929.13. Curtis, H. L., and Curtis, R. W., B.S.Jour. Research, vol. 12, p. 665, 1934.

Washington, March 15, 1934.

Calorimetric determination of the heats of...

Documents

Transcript of Calorimetric determination of the heats of...